AlexRa,
Since you were the one who wrote the newly introduced MySqlMigrations
class, which uses the MySqlScript class, can you please explain how we
can get around the issue that Frank is talking about? It looks like the
MySqlScript class has some hardcoded timeouts. Frank says that we might
be able to specify a timeout between 0 and 1 hour, but not more. Some DB
migrations, especially assets, might take more than 1 hour.
I'm really considering rolling back to using the old driver because this
one is causing too much grief and confusion at this critical time when
we want things to be stable.
Any light on this new driver, especially its limitations/bugs, and how
we can overcome them, would be very much appreciated so that we can try
to continue to use it in dev code.
Thanks.
Crista
On 6/18/2010 10:57 AM, Frank Nichols wrote:
The use of the MysqlScript class is at the heart of most of the
problems I have seen with the migration of mysql databases from 0.6.9
to 0.7.0. The problem seems to be that the default command timeout
value for processing a sql statement with this class is 30 seconds,
which is far to short to alter a table inserting a column when there
are hundreds of thousands of rows in that table.
In researching this it appears that the devart version of the new
MySql.data.dll has a version of the MysqlScript which allows you to
set the command timeout programmatically, however the mysql version of
this library does not have this property available. As far as I could
see there is currently no way to set the command timeout. So, it would
appear that automatic migrations are doomed to fail in any case where
they take more than 30 seconds.
Frank
On Fri, Jun 18, 2010 at 7:13 AM, Michael Cerquoni
<[email protected] <mailto:[email protected]>> wrote:
I have to say so far this new MySQL connector has been absolutely
nothing but trouble for Frank Nichols and myself testing Migration
from 0.6.9 to 0.7, All throughout the testing we have had nothing
but failure after failure from this new MySQL connector that has
been introduced. It has been nearly impossible to move forward on
to 0.7 because of this, I have no doubt that this will be a major
problem for everyone to migrate over if you have even what seems
like the most minimal amounts of data. At this time i really
have to give this new connector a -1. Can someone please explain
to me why the Original MySQL connector that seemed to work
perfectly for years has now been replaced with one that constantly
times out during migrations and gives nothing but trouble with
DNS, requires us to use additional connection string settings, and
basiclly makes moving on to 0.7 impossible for myself and
others?? I must be missing something here!
On Fri, Jun 18, 2010 at 6:06 AM, AlexRa <[email protected]
<mailto:[email protected]>> wrote:
AA> But I note that I has "pooling=false;" in my own string
though its
AA> not been in the example ini files for a while. Anyone
know what this
AA> does and if its wise to keep it in the string if I had a
working
AA> 0.6.9 system with it there before?
Hello Ai,
Can't tell you why the "pooling=false;" was there in the 1st
place, but
I've noticed that it did cause connection failures on a
Windows box during
performance testing (thousands mySQL requests in a loop, each with
connect/disconnect). The error didn't happen without the
"pooling=false;"
option. I don't know if that affects mySQL on Linux as well.
Best regards,
Alex
_______________________________________________
Opensim-dev mailing list
[email protected] <mailto:[email protected]>
https://lists.berlios.de/mailman/listinfo/opensim-dev
--
Michael Emory Cerquoni - Nebadon Izumi @ http://osgrid.org
_______________________________________________
Opensim-dev mailing list
[email protected] <mailto:[email protected]>
https://lists.berlios.de/mailman/listinfo/opensim-dev
_______________________________________________
Opensim-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/opensim-dev
_______________________________________________
Opensim-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/opensim-dev
